📸 ObjAI Scanner

📷 Screenshots

Home Screen	Details Screen

ObjAI Scanner is a modern Flutter application designed to identify real-world objects using Clarifai Vision AI, then fetch educational facts from Wikipedia, and store the result locally using Hive.

The project is built as a Clean Architecture example using Riverpod for scalable, testable state management.

---

✨ Features

• AI Object Identification
  Capture an image (camera or gallery) and send it to Clarifai to detect the primary object.

• Knowledge Retrieval
  Fetch a structured summary, facts, and an image from the Wikipedia API.

• Local History (Hive)
  All scans — including original image paths and extracted facts — are saved and persist locally.

• Local File Storage
  Saves the original image to the app documents directory.

• Clean Architecture
  Strict separation into Domain, Data, and Presentation layers.

• Riverpod State Management
  Compile-time safe and scalable.

---

🏗️ Tech Stack

Category	Technology	Purpose
Language	Dart / Flutter	Cross-platform mobile app
Architecture	Clean Architecture, Repository Pattern	Scalability & testability
State Management	Riverpod	Robust & compile-time safe
Local Storage	Hive + hive_flutter	Fast NoSQL persistence
Networking	dio	API calls (Clarifai + Wikipedia)
Utilities	image_picker, uuid, path_provider	Image selection & file handling

---

🧩 Architecture

lib/
├── core/
│   ├── storage_helper.dart
│   └── utils.dart
├── data/
│   ├── datasources/
│   │   ├── local_datasource.dart
│   │   ├── vision_datasource.dart
│   │   └── wiki_datasource.dart
│   ├── models/
│   │   ├── scan_model.dart
│   │   ├── scan_model.g.dart
│   │   └── wiki_result.dart
│   └── repositories/
│       └── scan_repository_impl.dart
├── domain/
│   ├── entities/
│   │   └── scan_entity.dart
│   └── repositories/
│       └── scan_repository.dart
└── presentation/
    ├── providers.dart
    └── screens/
        ├── home_screen.dart
        └── details_screen.dart

Layer Responsibilities

Domain

• Contains only pure business logic
• Defines:
  • ScanEntity
  • ScanRepository (abstract)

Data

• Implements repository logic
• Handles:
  • Clarifai API calls
  • Wikipedia API calls
  • Hive database CRUD
  • Mapping between Models and Entities

Presentation

• UI screens
• Riverpod providers & notifiers
• No business logic inside widgets

---

🔑 Setup & Installation

1️⃣ Clone the Repository

git clone https://github.com/mohamedchouat/objAi.git
cd objAi

2️⃣ Install Dependencies

flutter pub get

3️⃣ Configure Clarifai API Key

Open:

lib/presentation/providers.dart

Replace:

final clarifaiApiKeyProvider =
    Provider<String>((ref) => 'YOUR_CLARIFAI_API_KEY_HERE');

with your real key.

4️⃣ Generate Hive Adapters (if models changed)

flutter pub run build_runner build --delete-conflicting-outputs

5️⃣ Run the App

flutter run

---

📝 Usage

1. Open the app
2. Tap Camera or Gallery
3. Select an object (example: Cat, Eiffel Tower, Car, Motorcycle)
4. ObjAI will:
   • Identify object via Clarifai
   • Fetch summary and image via Wikipedia
   • Save scan result locally
5. View rich details in the Details Screen
6. History persists automatically

---

🗂️ Example Stored Entry (JSON)

{
  "id": "c3f1a820-1a2b-4f0a-bf01-442d88fc983e",
  "label": "Eiffel Tower",
  "summary": "The Eiffel Tower is a wrought-iron lattice tower in Paris...",
  "wikiImage": "https://upload.wikimedia.org/.../Eiffel_Tower.jpg",
  "localImagePath": "/data/user/0/com.objai/files/scan_12345.jpg",
  "date": "2025-01-04 18:22"
}

---

📦 Dependencies (pubspec excerpt)

dependencies:
  flutter:
    sdk: flutter
  dio: ^5.0.0
  riverpod: ^2.0.0
  flutter_riverpod: ^2.0.0
  hive: ^2.2.3
  hive_flutter: ^1.1.0
  path_provider: ^2.0.0
  image_picker: ^1.0.0
  uuid: ^4.0.0

dev_dependencies:
  build_runner: ^2.4.0
  hive_generator: ^2.0.0

---

🤝 Contributing

Pull requests are welcome!

Before submitting, please run:

dart format .

---

🛡️ License

This project is licensed under the MIT License.